Search

A hardware CPG model for biped gait control using pulse-type hardware neural networks(P-HNNs) composed of analog electronic circuits to generate time-series neural signal patterns for controlling human walking and running is reported. Human walking and running are controlled by central pattern generators located in a spinal cord. Apart from this spinal system, gait and gait speed is known to be changed by a signal from a higher center input to a brainstem-spinal cord projection. However, the specific circuit configuration of the neural circuits for controlling gait locomotion is still unknown. In recent studies, walking and running have been estimated to be controlled with five simple time-series pulse patterns generated by the spinal cord and the duration of their pulses. We have previously used P-HNNs to generate time-series neural signal pulse patterns for controlling human walking and running. In the model, the order of the time series was determined manually by trigger input. In addition, the change in the pulse pattern period when switching from walking to running was not reproduced. In this paper, we propose a hardware CPG model for biped gait control, with a network configuration capable of automatically determining the order of the time series, and changing the pulse pattern period. Based on the circuit configuration of the proposed CPG model, circuit simulations and CPG circuits fabricated with surface mount components were measured. The measured results confirm the automatic output of the neural signal patterns for controlling walking and running in time-series order. We also confirmed the pulse pattern period becomes shorter when switching from walking to running.

For constructing an isolated tooth identification system using deep learning, Igarashi et al. (2021) began constructing a learning model as basic research to identify the left and right mandibular first and second premolars. These teeth were chosen for analysis because they are difficult to identify from one another. The learning method itself was proven appropriate but presented low accuracy. Therefore, further improvement in the learning data should increase the accuracy of the model. The study objectives were to modify the learning data and increase the learning model accuracy for enabling the identification of isolated lower premolars.Static images of the occlusal surface of the premolars made from the dental plaster casts of dental students were used as the training, validation, and test data. A convolutional neural network with 32 hidden layers, AlexNet, convolutional architecture for fast feature embedding, and stochastic gradient descent was used to construct four learning models.The accuracy of the identification model increased using static images of the occlusal surface of the teeth with the adjacent teeth deleted as the training and validation data; however, a learning model that could perfectly identify the teeth could not be realized.Static images of the occlusal surface of the teeth with the adjacent teeth deleted should be used as both training and validation data. The ratio of the numbers of training, validation, and test data should be optimized.